English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Single-shot diffraction imaging of isolated nanosized particles has seenremarkable success in recent years, yielding in-situ measurements withultra-high spatial and temporal resolution. The progress ofhigh-repetition-rate sources for intense X-ray pulses has further enabledrecording datasets containing millions of diffraction images, which are neededfor structure determination of specimens with greater structural variety andfor dynamic experiments. The size of the datasets, however, represents amonumental problem for their analysis. Here, we present an automatized approachfor finding semantic similarities in coherent diffraction images withoutrelying on human expert labeling. By introducing the concept of projectionlearning, we extend self-supervised contrastive learning to the context ofcoherent diffraction imaging. As a result, we achieve a semantic dimensionalityreduction producing meaningful embeddings that align with the physicalintuition of an experienced human researcher. The method yields a substantialimprovement compared to previous approaches, paving the way toward real-timeand large-scale analysis of coherent diffraction experiments at X-rayfree-electron lasers.

Finding the semantic similarity in single-particle diffraction images using self-supervised contrastive projection learning